The Instrument You Don't Trust Decision Tree
Why this matters
A reading that does not make sense puts you at a fork: either the equipment is doing something strange, or your instrument is lying to you. Pick wrong and you waste an afternoon. Believe a faulty "dead" reading and you can get hurt. This tree walks you from the surprising reading to a verdict on whether to trust the number, fix the instrument, or stop and verify another way.
Start here: is this a safety-critical reading?
Before troubleshooting the instrument, ask what you are about to do with the number.
- If you are about to rely on a reading to declare something safe to touch, safe to open, or safe to work on (for example, an absence-of-energy reading): do not trust a single instrument that has surprised you. Verify on a known live source first, then on your target, then on a known live source again. This "live-dead-live" habit proves the instrument was working at the moment it mattered. Never declare something de-energized on a meter you have not just proven.
- If the reading is informational and not life-safety: continue down the tree and diagnose the instrument calmly.
Step 1: rule out the leads, battery, and settings
Most untrustworthy readings are user-level faults, not internal calibration. Check these first, in order.
- Battery. A weak battery makes many instruments read low or erratic. If marginal, replace it and retest.
- Leads and probes. Inspect for nicks, corrosion, and intermittent breaks. Wiggle-test them on a known source. A flickering reading while you flex a lead is a bad lead, not a bad system.
- Range and mode. Confirm you are on the correct setting for what you are measuring. A "wrong" number is often the right answer to the wrong question.
- Connection at the test point. Dirty or loose contact means you are measuring the connection, not the circuit.
If any of these was the problem, fix it and re-read. The instrument is fine.
Step 2: read a known value
If the basics check out and you still doubt the number, stop measuring the unknown and measure something known.
- Read a known reference: a fresh cell of known voltage, a precision resistor, a calibrated source, or another instrument you trust.
- If the instrument disagrees with the known value: the instrument is the problem. Tag it out of service and switch to a known-good one. Do not finish the diagnosis on a meter you just proved wrong.
- If the instrument agrees with the known value: the instrument is probably fine, and the surprising reading is real. Go to Step 4.
Step 3: break the tie with a third opinion
If two instruments disagree and you cannot tell which is right, you only know one is wrong.
- Bring in a third instrument or a third known reference.
- The odd one out is your suspect. Retire it until it can be checked against a traceable standard.
Step 4: the instrument is good, so believe the world
Once the instrument is proven, trust the reading and turn back to the equipment. Use observable reality as a sanity check.
- If the reading says the circuit is dead but the equipment is clearly energized or running: do not accept "dead." Re-verify with the live-dead-live method. An open lead or wrong setting can fake a dead reading, and that specific lie is dangerous.
- If the reading implies the system cannot work, yet it works: your method or test point is likely off. Re-check where and how you are measuring.
- If the reading is plausible and the instrument is proven: it is real. Diagnose the equipment, not the meter.
Quick reference: trust, fix, or stop
| Symptom | Most likely cause | Action |
|---|---|---|
| Reading flickers when leads move | Bad lead or connection | Replace lead, retest |
| Low or erratic across the board | Weak battery | Replace battery, retest |
| Disagrees with a known value | Instrument out of cal or faulty | Tag out, use known-good |
| Two meters disagree | One is wrong, unknown which | Third reference breaks tie |
| "Dead" but equipment is live | Open lead, wrong range, faked zero | Live-dead-live verify before trusting |
| Plausible and instrument proven | Reading is real | Diagnose the equipment |
The discipline that keeps you safe and sane
Two habits prevent almost every instrument-trust failure. First, prove the meter on a known value before you let it condemn a part or declare a circuit safe. Second, when a reading and the visible behavior of the equipment disagree, find out which one is lying before you act on either. A surprising number is a question, not an answer.
References
- OSHA guidance on verifying voltage testers on a known source (live-dead-live) before relying on a de-energized reading
- Manufacturer instrument documentation for self-checks, accuracy, and lead care
- NIST-traceable calibration practice for resolving instrument disagreement
- See related: Calibration: When Your Meter Lies; Stored Energy: Pressure and Charge Decision Tree